AWhat changed
Finding one relevant paragraph is different from asking what themes run through thousands of documents. GraphRAG extracts entities and relationships, groups them into communities and prepares summaries at several levels. A question can then draw on these summaries through a map-and-reduce process that combines partial answers. Edge, Trinh and colleagues demonstrate the approach on large text collections. The contribution is this graph-and-community summarization pipeline for collection-wide questions, separate from earlier retrieval, community-detection and language-model methods.
Scientific scope
Finding one relevant paragraph is different from asking what themes run through thousands of documents. GraphRAG extracts entities and relationships, groups them into communities and prepares summaries at several levels. A question can then draw on these summaries through a map-and-reduce process that combines partial answers. Edge, Trinh and colleagues demonstrate the approach on large text collections. The contribution is this graph-and-community summarization pipeline for collection-wide questions, separate from earlier retrieval, community-detection and language-model methods. The evaluation uses generated global questions and model-judged answer preferences, rather than a complete factual answer key or independent reader study. Gains vary by criterion and hierarchy level. Query-stage context savings exclude graph extraction and summary preparation. Entity links and summaries can be wrong. The original eight-author version is retained; later bylines are not retroactively substituted. This method is not the invention of all graph-based retrieval or a guarantee that a fluent summary preserves every source condition.
Evidence
The linked original works support the scoped statement and contribution attribution. WRS has not independently replayed experiments or reconstructed a full mathematical proof. The evaluation uses generated global questions and model-judged answer preferences, rather than a complete factual answer key or independent reader study. Gains vary by criterion and hierarchy level. Query-stage context savings exclude graph extraction and summary preparation. Entity links and summaries can be wrong. The original eight-author version is retained; later bylines are not retroactively substituted. This method is not the invention of all graph-based retrieval or a guarantee that a fluent summary preserves every source condition.
Contributors
- The named authors in the research record · SCIENCE: Source-listed human researchers and engineers for the exact method or resource. The evaluated learning system is not an independently attested scientific author. Earlier methods, funding and data-collection roles remain separate. Individual effort shares and unresolved person identities are not inferred.
Provisional WRS assessment, edition 2026-10-23.9. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction